6.0 Marine Sediment Remediation

6.1 Remediation Requirements

• sediment-dwelling organisms based on published data;

• human consumers of shell fish and fish which may bioaccumulate contaminants through the food chain; and

• species at higher trophic levels (seafood-eating birds) exposed at relatively high concentrations due to bioaccumulation within the food chain.

For DDT, the acceptable concentration in seafood food for human consumption was taken to be 0.1 mg/kg. A bioaccumulation factor (BCF) of 10 was assumed (Egis, 2001), based on measurement in mud snails (amphibola crenata), giving an SAC of 0.01 mg/kg. A check against higher trophic levels was made using literature bioaccumulation factors.

A similar approach was adopted for dieldrin. Again the starting point was the food standard of 0.1 mg/kg. The same BCF of 10 was used based on sediment and mud snail concentrations, resulting in the SAC of 0.01 mg/kg. A check that this would be protective of birds was carried out.

An SAC of 0.01 mg/kg for both DDT (applied as DDX) and dieldrin (applied to include aldrin and lindane) will be protective for recreational users of the foreshore, e.g. children playing. While children playing in mud can ingest larger quantities of sediment than for the residential situation and dermal exposure is expected to be greater (a child getting covered head to toe in mud), the frequency of exposure is expected to be less than the residential exposure frequency. The combination of these two factors will at worst be about even but probably favour a lower daily average sediment intake for the recreational scenario over the residential scenario. Given the marine SACs are at least two orders of magnitude lower than the corresponding human-health guidelines, a large margin of safety is provided for the recreational user.

The other SACs were the ISQG-Low marine sediment guidelines from ANZECC (2000). ANZECC (2000) recommends adjusting the guidelines for total organic carbon (TOC) as the guideline increases with increases in TOC. Measuring TOC is important for non-ionic organic compounds such as DDX and ADL, but the consent did not specify this and it was never carried out. Particle size distributions would also be helpful to give context to differences in contaminant concentrations between sample locations. The particle size distribution also affects the suitability of a sediment habitat for various biota.

The consents specified the remediation method in a relatively detailed way.

Condition 10(j) of both RM030521-General and RM030522-Coastal Marine required as part of a suite of procedures:

A coastal marine excavation works procedure specifying but not limited to:

1. Excavation depth which shall not be limited to depth by distance but shall be such depth as is required to enable the Consent Holder to meet the validation standards required under the Soil Acceptance Criteria (Marine Sediments category);

2. Unless prevented by reason of practical considerations excavation in the Coastal Marine Area shall take place by way of long reach excavator;

3. As far as practicable, works will be completed in one tidal cycle, the area excavated covered with a layer of gravel to prevent the migration of fines onto the excavated surface. Works will recommence on the next low tide cycle and be carried out in the same manner;

4. In the event the validation samples do not establish that the criteria required by the Council has been achieved then the area shall be re-excavated to the depth that the remediation levels are met;

5. A similar method will be adopted for removal of sediments up to 100m offshore adjacent to the FCC Landfill, with a gravel access being constructed to the limit of the works. The excavator will then, working on low tide, excavate the sediments and retreat toward the FCC landfill, recovering gravels as areas are completed;

The coastal marine consent further required monitoring under Condition 19:

Monitoring

19. The Consent Holder shall undertake a programme of macro invertebrate and sediment quality sampling at selected sites within and distant to the area of sediment excavation. The monitoring programme shall be approved by the Council’s Compliance Coordinator and shall be undertaken before works authorised under this consent commence, (unless Council holds sufficient prior sampling records) and at 12 months, 24 months and 36 months following completion of the works undertaken pursuant to this Consent.

and imposed further constraints on the timing of the remediation in conditions 20 and 21, which specified:

Timing of Excavation

20. The marine sediment excavation process shall take place as late in the remediation process as practicable.

21. Excavation of marine sediments shall not be carried out during rain or under water. The excavated area will be covered with clean fill appropriate for the marine environment prior to being submerged by the tide.

It is now a matter of record that the marine sediment remediation on both the east and western sides of the site failed to meet the SACs by a large margin, although there is some evidence to suggest that the situation has improved with time. The question is now whether the remediation was carried out to the extent practicable. This is a subjective test dependent on a number of factors.

The circumstances of the two marine sites are different and must be examined in turn, however, a common working definition of the extent practicable is required.

6.2 Working Definition of Remediation to the Extent Practicable

As noted, the extent practicable is a subjective test. It depends on such things as:

• constraints imposed by the immediate physical environment, not least, in this case, the tidal nature of the site imposing time constraints and the difficulty working in mud;

• constraints imposed by the amount of money available (with contractual arrangements presumably imposing additional constraints) – virtually anything is possible if enough time and money is available, but there never is;

• the condition of the wider environment – SACs will not be achievable if adjacent areas have higher concentration within sediment that is mobile;

• constraints imposed by the consent conditions in terms of methods and timing; and

• perceptions as to whether residual effects are acceptable or adverse, and if the latter, weighing up the potential benefits of further remediation against potential harm of the remediation itself (temporary destruction and/or permanent change to the habitat) and the cost.

For this assessment it is assumed that methods reasonably similar to those specified in the consent had to be used. More sophisticated (but not necessarily more effective given the circumstances) methods could have been adopted, but the cost would have been much greater.

It is also assumed that the remediation was to be limited to the low tide mark at the FCE East beach (to go into the channel increases the difficulty markedly).

It is further assumed that it was not physically or economically practical to remediate a considerable part of the western bay between Grossi Point and the FCC Landfill vicinity.

As with other aspects of the extent practicable definition, determining what effects on receptors are acceptable is a subjective decision. Essentially it is necessary to balance often intangible benefits against equally intangible perceptions of harm.

In making the final judgement it is necessary to have an understanding of why the remediation was not as successful as intended (which is the case). This is required in order to determine whether, for example, a better application of the method or a modest modification to the extent, might have achieved a better outcome, or alternatively, no matter what was done (within reason), the intended remediation was inherently flawed. If the reason for failure cannot be worked out, then this judgement cannot be arrived at.

6.3 East Marine Sediment Remediation

6.3.1 General

This section is based on information presented in the Validation Report (SKM, 2008) and a number of other sources as outlined in Section 4. A brief description of the remediation works has been included here as an understanding of this is required to interpret the validation data and a detailed description was not included in the Validation Report.

Figure 4 shows the approximate layout of the remediation excavation in relation to the bund / sea wall. This figure is based on Figure 41 in the Validation Report (SKM, 2008). No scale or lateral reference system was provided in the original figure making it difficult to determine what part of the bund the horizontal line at the top of the drawing represents (we have assumed it represents the historical ‘top of bund’ based on a review of site records). Consequently, the locations of the samples cannot be accurately determined from the information presented. The approximate scale shown on Figure 4 has been estimated from sample position information obtained separatel

6.3.2 Pre-excavation testing

In April 2005, a series of sediment samples were collected to attempt to characterise the contamination prior to remediation. The sample locations are shown in green on Figure 4. Samples were collected variously from 0.15, 0.5 and 1.0 m depths and analysed for DDX and ADL. The results indicated elevated concentrations of these contaminants, with only three of the 57 samples analysed for DDX returning concentrations below the marine SAC of 0.01 mg/kg. The maximum DDX concentration of approximately 8.5 mg/kg was detected at 0.5 m depth close to the toe of the former bund. There were approximately 19 samples with DDX concentrations above 1 mg/kg, with most of these located along the toe of the former bund. Concentrations above 1 mg/kg were detected at 0.5 m depth at six locations, with no analytical data from below that depth to show the vertical extent of the contamination.

The maximum ADL concentration detected was 0.65 mg/kg and 22 of the 57 samples returned ADL concentrations below the marine SAC of 0.01 mg/kg.

A number of other investigations have tested sediment samples from the wider vicinity along the eastern foreshore. These results are discussed further in Section 1.7.

6.3.3 Remediation Excavation

The removal of marine sediment occurred in early May 2005. The excavation was undertaken in a series of individual cells (FS1 – FS14 on Figure 4), extending into the channel by up to about 30 m from the toe of the bund. As required by the consent, each cell was excavated and backfilled within one tidal cycle i.e. above water. The depth of excavation ranged from approximately 0.25 m to 1.0 m below the original surface, with an average of about 0.5 m. This was greater than originally planned. The excavations were backfilled with an imported coarse river sand/gravel (see Section 1.5.6 for further detail).

The remediation of the area around the former stormwater surge chamber (see Section ) was completed in July 2005, shortly after the marine sediment remediation. Figure 4 shows the approximate location of the surge chamber in relation to the sediment excavations.

From the information obtained, it is not clear whether the area represented by the pre-excavation samples 1072 – 1089 has been remediated (Figure 4). It would appear that these areas were not included in the marine sediment remediation works i.e. they are not within cells FS1 – FS14. However, it is likely that at least the central part of this strip of beach was remediated during the works to remove the surge chamber (subgrades SG10 and SG13 on Figure 2). In addition, the works to construct the sea wall are likely to have overlapped these sample locations to some extent. This is difficult to confirm as the position of the samples cannot be accurately determined from the location plan.

A review of site photos shows that the backfilled area was ‘root-raked’ in June 2005, the month after the remediation was completed. No other mention of this has been seen in any other documentation reviewed. It appears that the entire remedial excavation was dragged through with a comb-like bucket on the excavator. The tines on the bucket appear to be over 0.5 m in length. It is not known why this was undertaken, and whether the full depths of the tines were used in the process. If this was the case, it would have had the effect of mixing the backfill and potentially mixing in underlying in-situ material, given the length of bucket tines.

Although there are some uncertainties in the remediation work methods and the subsequent validation data, these do not significantly affect the overall conclusions regarding the marine sediment remediation.

6.3.4 Validation Sampling

Validation samples were collected from the base of each excavation immediately prior to backfilling. These samples are presented in black on Figure 4 and labelled as ‘Layer 2 12th May 2005’ on the legend. All 86 of these samples were analysed for the full OCP suite, including DDX and ADL. A further 31 samples were analysed for TPH, OPP, ONP and VOCs. Six samples were analysed for the 10% suite which includes heavy metals, PAHs and PCBs. No samples were collected from the walls of the excavation, so the conditions at the edge of the excavation cannot be determined.

A subsequent set of 16 samples was collected about three months after the remediation works. These samples were collected from approximately 0.3 m depth in a line along the beach, as indicated by the red samples on Figure 4, and analysed for the full OCP suite. It is not stated what the target stratum was for this sampling, however, the sample depth would tend to imply that the samples were collected from within the remediation backfill material.

A further 15 samples were collected from the East beach by TDC and CH2M HILL in May 2007, approximately two years after the beach remediation was completed. Three of the samples were collected from 0.15 m below the surface and the remainder were collected between 0 and 0.05 m. The surface samples were collected over four three-part transects down the beach, spaced out along the beach. Two of the deeper samples were collected centrally on the beach and the other was collected opposite the first residential property south of the site. All samples were analysed for OCPs, ammonia and nitrate.

6.3.5 Compliance with SAC

The validation results indicate that the remediation of the eastern marine sediments did not achieve compliance with the SAC and sediment remaining in the area had concentrations significantly above the target values. The exceedances predominantly relate to DDX and ADL. In addition, the post remediation sampling of the backfill material indicated that the coarse sand and gravel backfill had been re-contaminated in the three months after the remediation had been completed. Details of the exceedances are discussed below and the significance of the exceedances is assessed in Section 1.9.

The Validation Report (SKM, 2008) combines the results from samples taken from the excavation base with post-remediation sampling of the backfill. This is not appropriate as the two datasets represent different situations and are likely to be from different populations. The datasets have been discussed separately below.

Initial Validation Sampling

Of the 86 samples taken from the excavation base, 69 had contaminant concentrations above the SAC. The majority of these exceedances related to elevated DDX concentrations, with 65 samples above the SAC of 0.01 mg/kg for DDX. The maximum DDX concentrations of 125 mg/kg and 58 mg/kg were detected in cell FS11, close to the surge chamber discharge point. All other DDX concentrations were below 3 mg/kg.

ADL concentrations above the SAC of 0.01 mg/kg were detected in 23 of the 86 samples, with a maximum concentration of 3.9 mg/kg (also in cell FS11).

As noted in the assessment of the marine sediments as residential backfill (Section 1.1.4.2), the Validation Report incorrectly assumed the dataset was log-normally distributed. However, given the 95% UCL estimates were well above the SACs (at least five times for ADL and two orders of magnitude for DDX) regardless of method, this error is of no particular significance.

The Validation Report (SKM, 2008) stated that six nickel results were above the SAC, with a maximum concentration of 80 mg/kg. However, the report incorrectly used a superseded SAC of 21 mg/kg instead of the correct value of 70 mg/kg. The minor exceedance in nickel concentrations is not significant.

The detection limit for the PCB analysis (0.03 mg/kg) was not low enough to assess the data adequately against the SAC of 0.023 mg/kg. Five of the six samples analysed for PCB were below the detection limit, with a concentration of 0.07 mg/kg detected in a sample from FS10 (again, close to the surge chamber).

A similar situation exists for chlordane, where the laboratory detection limits (variable) were typically not low enough for comparison with the SAC of 0.0005 mg/kg. However, chlordane was detected above the laboratory detection limits in ten samples, with a maximum concentration of 0.16 mg/kg.

These discrepancies are not significant given the large non-compliance with the DDX SAC.

Post Remediation Sampling (May 2005)

Of the 16 samples post-remediation samples collected, 15 had DDX and ADL concentrations above the respective SAC. The DDX concentrations ranged between 0.4 and 6.4 mg/kg and the ADL concentrations between 0.02 and 0.28 mg/kg. As discussed above, it is not entirely clear what these samples represent. However, assuming that they were taken from within the sand/gravel backfill material, the samples are likely to represent fine sediment that has infiltrated the coarse backfill material. The potential source of this contamination and its significance are discussed in Section 1.9.

CH2M-HILL / TDC samples (May 2007)

All of the 12 surface (0 -0.05 m depth) samples had DDX concentrations above the SAC of 0.01 mg/kg, with concentrations ranging between 0.05 and 1 mg/kg. Four of the ADL concentrations in the same samples were above the SAC, with a maximum concentration of 0.05 mg/kg.

The three samples taken from 0.15 m depth had higher DDX and ADL concentrations. The DDX concentrations ranged from 3.2 to 27.5 mg/kg, with the maximum concentration detected in a sample from close to the base of the sea wall, at the top of the beach. The ADL concentrations in the same samples ranged from 0.10 to 0.66 mg/kg.

Ammonia and nitrate concentrations in all samples were below the corresponding laboratory detection limits of 5 and 1 mg/kg respectively.

6.3.6 East Marine Backfill

A total of approximately 5,000 m3 of sand/gravel was imported for use as backfill for the marine excavations. Approximately 2,900 m3 of this was used for the east marine excavation and beneath the eastern bund / sea wall. The material was sourced from four different locations and at least two samples were taken from each location. The material was described a well-graded sandy gravel (MWH, 2009g), although no particle size distribution test data was available. The volume and final destination for each source is not known.

Ten samples were taken from the marine backfill material and analysed for the full OCP suite. This represents one sample per 500 m3, which complies with the RAP requirements. Two of these samples (20%) were also analysed for a metals suite and a single sample was analysed for PAHs.

One of the 10 samples had a DDX concentration of 0.012 mg/kg, slightly above the marine SAC of 0.01 mg/kg. This exceedance is not significant, particularly as the result is well below the DDX concentrations remaining at the marine excavation extents. All other DDX and ADL concentrations were below the SAC.

One of the two samples analysed for nickel returned a concentration (165 mg/kg) above the SAC of 70 mg/kg for nickel. The average of the two samples was 102 mg/kg. It would have been desirable to have taken more samples for nickel to confirm the material as a whole complied with the SAC

The laboratory detection limit of 0.001 mg/kg for chlordane was not low enough to enable a comparison with the corresponding SAC.

Overall, the material imported as marine backfill sufficiently complies with the SACs.

6.4 West Marine Sediment Remediation

6.4.1 West Marine Excavation

This section is based on information presented in the Validation Report (SKM, 2008) and a number of other sources as outlined in Section 4. As with the east marine remediation, a brief description of the remediation works has been included here as an understanding of this is required to interpret the validation data. Figure 5 shows the approximate layout of the remediation excavation in relation to the 15 m characterisation grid. This figure is based on information supplied by MWH.

6.4.2 Pre-excavation Testing and Remediation Approach

Various investigations of the sediment quality in the creek and foreshore area had shown elevated concentrations of DDX and ADL compounds. The results indicated that elevated concentrations extended over 100 m from the site boundaries into the Waimea Inlet. DDX concentrations across the broader estuary appeared to range from 0.1 to 0.9 mg/kg (GHD, 2006), well above the SAC of 0.01 mg/kg. Based on the experience gained during the remediation of the eastern foreshore, the viability of achieving compliance with the SACs in the west marine remediation was discussed between the Site Auditor, TDC and MfE. It was agreed that a more practical approach would be to attempt to remove the bulk of the contamination rather than adhering strictly to the SACs (GHD, 2006).

It is not clear how success was then to be judged as revised SACs were not developed. An initial suggestion of adopting a 1 mg/kg criterion for DDX was rejected by the Site Auditor on the basis that (GHD, 2006):

… it would appear that concentrations across the broader estuary typically range from 0.1 to 0.9 ppm. On this basis, it would be difficult to accept a revised criterion as high as 1 ppm.

Presumably meaning that choosing a criterion similar to existing concentrations was hardly an improvement. The Site Auditor went on to say:

In addition, given the need to remove contaminated sediments, collect validation samples and backfill with clean material within one tidal cycle, it will not be confirmed (from the laboratory results) that the validation samples will meet the adopted criterion until well after the clean up works have been completed.

A more practical approach would be to base the clean up on the principle of removing the bulk of the contamination sediments, and determining the extent of excavation on the basis of removing the highest concentration material. This would be consistent with the approach being taken for cleaning up the land at the site, where the objective has been to achieve a destruction efficiency of at least 90%, and would significantly reduce the time over which the target criterion will be reached through natural degradative processes. This would draw on the concept of “Clean Up to the Extent Practicable” … where it is not practicable to comply with very stringent criteria and the risk associated with the residual contamination is not excessive.

This is a sensible approach but inevitably relies on a somewhat subjective assessment in the absence of good information on the likely risks. The Site Auditor arrived at a preliminary estimate of the reduction in contaminant load for a revised excavation area that would achieve a 90% reduction in the contaminant load. It is not clear from the information reviewed whether this was to be a reduction in concentration (to, say, 0.06 – 0.09 mg/kg based on the measurements cited above) or a reduction in total mass within the area. It is also not clear over what total area the 90% reduction was being judged against, i.e. the immediate area of the beach or the complete bay bounded by Grossi Point.

6.4.2.1 Remediation Excavation

The initial remediation of the west foreshore area was completed at the beginning of May 2006. The excavation included three distinct areas: the creek; a swale from the end of the creek extending out into the inlet; and the general foreshore area. The swale was a drainage pathway, cut through the marine sediments by the discharge from the creek.

The remediation excavation of the creek removed at least 200 mm of soil from the walls of the creek and up to 600 mm from the base (EMS, 2006). Prior to excavation, vegetation was removed from the edges of the creek. Where the vegetation was from areas of low contamination, this was put to one side for re-use. Alternatively, the vegetation was taken off-site for disposal. The excavation of the swale and the foreshore removed between 400 and 600 mm of sediment.

The excavation was completed in a series of cells, each within one tidal cycle i.e. above water. Imported sandy gravel was placed in the completed excavations.

In September and October of 2007, the surface 100 mm layer of sediment was removed across the approximate area of the original foreshore remediation excavation as a follow-up to finding excessive concentrations in sediment samples in May and July 2007 (TDC, 2007a, 2008a). This excavation created a shallow pond on the eastern half of the foreshore. Similarly, approximately 100 mm of sediment was removed from the stream bed in cells O1, N1 and M2 (see Figure 5).

In mid-October 2007, heavy rainfall caused a discharge of sediment from treated fines stored on the adjacent landfill site (TDC, 2008). The discharge was visually apparent. In November 2007, TDC sampled three locations across the beach to determine the effects of the discharge. The results indicated that eastern end of the beach had been impacted with DDX concentrations between 2 and 6 mg/kg.

A 100 mm layer of imported sand/gravel was placed across this area in November 2007. This occurred after FCC Landfill had been covered with the capping layer of residential quality soil. No further sediment was removed.

6.4.3 Validation Sampling and Compliance with SAC

Overall, the scope of validation sampling for the west marine sediment remediation is reasonable for characterising the quality of the remaining sediments. The separate datasets are discussed below.

The validation results indicate that the remediation of the western marine sediments did not achieve compliance with the SAC and sediment remaining in the area had concentrations significantly above the target values. As with the eastern foreshore, the exceedances predominantly relate to DDX and ADL. The most elevated concentrations were detected in samples from the creek excavation.

Initial Remediation

A total of 146 validation samples were collected from the base of the excavation immediately prior to backfilling. All of these samples were analysed for the full OCP suite, including DDX and ADL. A further 82 samples were analysed for TPH, OPP, ONP and VOCs. Approximately 25 samples were analysed for the 10% suite (see Section 5.2) which includes heavy metals, PAHs and PCBs.

Of the 146 samples taken from the initial excavation extents, 94 had contaminant concentrations above the SAC. The majority of these exceedances related to elevated DDX concentrations, with 91 samples (97%) above the SAC of 0.01 mg/kg for DDX. The majority of the peak DDX concentrations were detected in samples from the creek excavation. Approximately 15 samples recorded DDX concentrations above 1 mg/kg (100 times the SAC) in this area, with a maximum concentration of 82 mg/kg detected in a sample from the north-west bank of the creek. The next highest DDX concentration in the creek area was 37 mg/kg. The only other DDX concentrations above 1 mg/kg were recorded in two samples at the eastern end of the beach, close to the high tide mark (9 – 15 mg/kg). It is understood that the sediment associated with these two elevated results was removed by EMS (Jenny Easton, TDC, pers. comm.), although no records have been located to confirm this.

The DDX concentrations in samples from the base of the foreshore excavation were typically below the SAC, indicating the full vertical extent of the contamination had been removed across the area. Samples from the edge of the foreshore excavation had DDX concentrations ranging from 0.01 to 0.13 mg/kg (excluding the hotspot at the eastern end of the beach). The samples from the base of the swale all returned DDX concentrations above the SAC of 0.01 mg/kg, with a maximum of 0.9 mg/kg.

ADL concentrations above the SAC of 0.01 mg/kg were detected in 19 of the 146 samples. Thirteen of the 19 exceedances were from the creek excavation, although the maximum concentration of 0.224 was detected at the eastern end of the beach (at the same location where elevated DDX concentrations were detected). As noted above, it is understood that the sediment associated with the peak concentration was removed.

The Validation Report (SKM, 2008) stated that nine nickel results were above the SAC, with a maximum concentration of 71 mg/kg. However, the report incorrectly used a superseded SAC of 21 mg/kg instead of the correct value of 70 mg/kg. The nickel concentrations are not significant.

The detection limit for the chlordane analysis (0.002 mg/kg) was not low enough to assess the data adequately against the SAC of 0.0005 mg/kg. Chlordane was detected above the laboratory detection limits (and the SAC) in a single sample, with a concentration of 0.083 mg/kg.

CH2M HILL / TDC Sampling

A total of 16 sediment samples were collected from the creek and western foreshore areas during the CH2M HILL and TDC sampling in May 2007 (see Section 1.5.4), approximately one year after the remediation. Fourteen of these samples were from the surface (0 – 0.05 m), with the remaining two from 0.15 m depth. The surface sampling was likely to have been representative of fine sediment that had migrated into the remediated area over the period since the remediation was carried out.

The DDX concentrations in all 14 surface samples were above the SAC of 0.01 mg/kg, with concentrations ranging from 0.05 to 7.4 mg/kg. The concentrations in almost all samples were above 1 mg/kg. Similarly high concentrations were present in the row of samples at the top of the beach and the row 15 – 20 m off-shore, with no particular pattern. The exceptions to samples with concentrations above 1 mg/kg were the two upstream-most creek samples (0.05 and 0.09 mg/kg) and a single swale sample (0.4 mg/kg). The ADL concentrations in the 14 surface samples were also above the SAC, with concentrations ranging from just above the criterion of 0.01 mg/kg up to 0.15 mg/kg.

The DDX concentrations in the two deeper samples were above the SAC (0.2 – 0.4 mg/kg), albeit slightly lower than the typical surface concentrations. The ADL concentrations in these samples were 0.004 mg/kg (below the SAC) and 0.014 mg/kg.

Ammonia and nitrate concentrations in all samples were below the corresponding laboratory detection limits of 5 and 1 mg/kg respectively.

13 July 2007

Three samples of sediment were collected from the west beach foreshore on 13 July 2007 (approximately one year after the initial remediation), although the exact location of these is unknown. The samples were analysed for the full OCP suite and a reduced metals suite. It appears that these and the CH2M HILL samples were used as the basis for undertaking the additional (September/October 2007) remedial excavation, although this could not be confirmed from the information reviewed.

All three samples had DDX and ADL concentrations above the SAC. The DDX concentrations ranged from 5 mg/kg to 24 mg/kg and the ADL concentrations ranged from 0.09 mg/kg to 0.49 mg/kg. It is not known at what depth these samples were taken or whether the sediment they represent was subsequently removed by the additional excavation.

September / October 2007 Remediation

A total of 14 samples were collected to validate the additional sediment removal in September / October 2007 and analysed for DDX and ADL.

Of the 14 samples analysed, 11 had DDX concentrations above the SAC with concentrations ranging between just over 0.01 mg/kg and a maximum of 0.59 mg/kg. ADL concentrations slightly above the SAC were detected in five samples, with concentrations ranging between 0.01 mg/kg and 0.02 mg/kg. It is not entirely clear what these samples represent. However, it is assumed that they were taken from the surface of the sediment remaining at the base of the additional remedial excavation.

November 2007 TDC Sampling

Three samples of sediment were collected by TDC from the west beach foreshore on 6 November 2007 and analysed for OCPs.

The three samples across the beach contained DDX concentrations ranging from 0.72 mg/kg (at the western end of the beach) to 5.9 mg/kg. The ADL concentrations ranged from 0.019 to 0.18 mg/kg. This appeared to confirm the theory that the sediment quality had been impacted by site runoff. The eastern end of the beach, where the more elevated concentrations were detected, was covered with 100 mm of imported gravel on 29 November 2007 to prevent mobilisation of the contamination. No sediment was removed.

6.5 Additional Sediment and Biota Monitoring Data

Tasman District Council has carried out monitoring of sediment quality and snails since the marine sediment remediation was completed (TDC, 2009b) to fulfil the monitoring condition of the consent. A baseline monitoring was carried out prior to the remediation in 2005. The sampling was conducted to determine whether the remediation resulted in an improvement in the quality of the marine habitat with respect to contamination levels. An earlier study considered snails to be the most appropriate bio-indicator of the success of the remediation (Landcare Research, 2002). This study included measurement of OCP accumulation in other fauna such as cockles, crabs and oysters.

In February 2005, TDC obtained sediment samples and collected snails from a series of locations on the western and eastern foreshores to benchmark conditions before the marine sediment remediation works occurred. A control site to the west of the site was also sampled to assess likely background concentrations in the site vicinity. The sediment samples and snail flesh were analysed for DDX and dieldrin and lindane compounds.

Three annual monitoring rounds have been completed (at similar sampling locations) since the majority of the foreshore remediation works were finished in 2006. In the west, three sediment samples have been collected at 40 m intervals along the swale, from close to the south-west corner of FCC Landfill out into the Waimea Estuary. Snails were collected from a 10 m by 10 m area approximately 40 m from the high tide mark in the central portion of the western beach. A sample of sediment was also collected at this location.

In the east, sediment samples were collected along a transect from the base of the sea wall out into Mapua Channel. The ‘top’ sample was from adjacent to a seep at the base of the wall, with samples collected at 8 m and 15 m intervals from that point. The sediment samples were collected from the top 1-2 cm of the sediment surface.

The initial and three subsequent set of results for the eastern and western beaches are shown in Table 3 and Table 4, respectively.

The latest monitoring results (TDC, 2009b) indicate a decreasing trend in OCP concentrations in both sediment and snails, although the small sample size is not definitive. The DDX concentrations in snail flesh have reduced on both east and west sites (noting that the 2005 snail results for East were a different species of snail to that now being sampled, with the original snail bioaccumulating more). In the west, the DDX concentration in snail flesh has dropped from 51 mg/kg in 2007 to 3.5 mg/kg in February 2009. The corresponding dieldrin concentrations in the west have dropped from 2.2 mg/kg in 2007 to 0.2 mg/kg in February 2009, a ten fold decrease. In the east DDX concentrations in the snail flesh have dropped from 0.54 mg/kg to 0.03 mg/kg, an 18-fold decrease. The lesser bioaccumulation in the east versus the west is due to the different species of snail being tested (top shell versus mud snail).

Every sediment sample site has also shown a reduction in OCP concentrations when compared to the previous results For example, in the west, the DDX concentrations along the swale have reduced from between 0.4 and 3.9 mg/kg to between 0.1 and 0.6 mg/kg. The DDX concentration at the west snail sampling location has reduced from 16.6 mg/kg to 0.23 mg/kg over the three monitoring rounds. The dieldrin and lindane concentrations have also decreased at each site.

Bioaccumulation has possibly increased for both DDX and dieldrin since 2007, perhaps reflecting increased snail age, with bioaccumulation factors now higher than the 2005 samples, albeit at much reduced concentrations. This may be an effect of increased snail lifespan due to the lower concentrations than pre-remediation, allowing the snails to bioaccumulate relatively more.

In the east, the ‘top’ sediment sample is from adjacent to a groundwater seep. The DDX concentrations have decreased from 2.4 mg/kg to 0.03 mg/kg over the three monitoring rounds. In earlier monitoring rounds, there had been some algal growth on the beaches, particularly around the seepage area. In the latest monitoring round ‘no excessive algal growth’ was noted.

6.6 Off-site Sediment Data

A number of investigations have involved testing sediment samples from locations away from the two main remediation areas.

On two occasions, sediment samples were collected from beneath the wharf approximately 40 m to the north-east of the site. The sampling events in 2005 and 2007 indicated that elevated OCP concentrations are present in the sediment beneath the wharf to depths of at least 0.4 m below the surface. For example, DDX concentrations ranging between about 1 mg/kg and 60 mg/kg were detected in samples from various depths. The ADL concentrations were much lower. The sediment at this location is covered by a layer of gravel armouring, a reflection of the high energy tidal environment which has removed the fine material leaving the larger stones.

A figure showing what appear to be DDX results from sediment samples in the wider site area was found during a review of the EMS site files. The date of sampling was not provided on the figure, although the document was tabled in a peer review panel meeting in December 2005 (Jenny Easton, TDC, pers. comm.). However, the results show sediment adjacent to the Mapua Channel beyond the wharf, approximately 100 m north north-east, also had elevated DDX concentrations (1 and 17 mg/kg). DDX concentrations in samples further along Mapua Channel appeared to decrease significantly. Concentrations were below the laboratory detection limit of 0.01 mg/kg in samples taken from approximately 500 and 800m to the north north-east of the site. A sample from the tip of Grossi Point, approximately 400 m to the south, returned a DDX concentration of 0.05 mg/kg. Two samples from 50 to 100 m west of the site returned DDX concentrations of approximately 0.01 mg/kg.

In October 2007, TDC collected a series of surface sediment samples from the Waimea Estuary in two transects. This included samples up to 250 m to the south of the western foreshore, within the tidal mud-flats (TDC, 2008a). The samples represented the top 0.1 m of the sediment.

The results are shown in Table 5. DDX concentrations ranged between 0.07 mg/kg and 0.24 mg/kg, with an average of about 0.15 mg/kg, not dissimilar to the most recent of the annual sediment samplings (Table 4). The sample collected at approximately 250 m from the foreshore had a DDX concentration of 0.13 mg/kg, showing how far from the shore similar to average concentrations can be. Dieldrin concentrations ranged from 0.0016 to 0.0061 mg/kg, with an average of 0.003.

Concentrations of DDX in snail flesh ranged from about 4 to 73 mg/kg, with the two highest concentrations occurring 85 and 167 m from the beach. These concentrations are much higher than that measured in TDC’s annual snail surveys on the beach, although the highest snail concentrations are similar to the 2005 baseline measurements. Bioaccumulation factors are generally much higher than the annual samples, ranging from 30 to 400 for DDX and 70 to 950 for dieldrin. Again it might be result of snails living longer at the lower concentrations and therefore bioaccumulating to similar concentrations as snails living on sediments with higher concentrations. Toxic effects causing mortality may be defining the upper limit of the concentration in the snails.

A number of other earlier investigations collected sediment samples from the wider area of Mapua Channel and Waimea Inlet. These included a set of samples collected by TDC in 1996 which showed the following general patterns of contamination (T&T, 2003a);

• DDX contamination in shallow (0 – 0.25 m) sediment to the south of the eastern remedial excavation was variable but appeared to reduce with distance from the site. A concentration of 1.5 mg/kg was detected about 20 m south of the site, but concentrations were typically close to or below the SAC of 0.01 mg/kg beyond about 50 m from the site;

• in the Waimea Inlet, samples from up to 200 m south of the foreshore showed a similar pattern of contamination to that indicated in the 2007 TDC sampling discussed above. For example, a DDX concentration of 0.9 mg/kg was detected in shallow sediment (0 -0.25 m) approximately 100 m from the foreshore; and

• ADL concentrations in shallow sediment away from the site (for both east and west) were typically below the SAC of 0.01 mg/kg.

6.7 Significance of Residual Marine Sediment Contamination

6.7.1 Eastern Foreshore

6.7.1.1 Mass of contaminant removed

It is possible to estimate the total mass of DDX compounds removed by using the product of the approximate volume of sediment removed and an estimate of the average DDX concentrations of that material. Using the available data, it is estimated that the mass of DDX compounds removed was in the order of 8 kg. There is not enough data to estimate the mass of DDX contamination remaining in sediment outside the east marine excavation in the Mapua Channel. Consequently, it is not possible to estimate the proportional mass reduction achieved for the sediment remediation in the east.

6.7.1.2 East Current Status

The sediment quality on the eastern foreshore has not been fully characterised. However, it is clear that elevated concentrations of OCP compounds remain in sediment in the site vicinity. Post-remediation sampling has confirmed concentrations of DDX more than 100 times the SAC (i.e. above 1 mg/kg) remain at some locations in the vicinity of the remedial excavation (a maximum concentration of 125 mg/kg was detected in a sample from the base of the remedial excavation, although this was not typical). Elevated DDX concentrations are also present in shallow sediment up to 100 m to the north of the site (including beneath the adjacent wharf), with concentrations of between 1 and 60 mg/kg detected.

The ADL concentrations on the eastern foreshore are much lower than the DDX concentrations, but still elevated. The post remediation sampling showed that most ADL concentrations were below 1 mg/kg, but concentrations above 0.1 mg/kg were not uncommon.

Although there are high concentrations of nutrients in groundwater discharging from the site, concentrations of nitrogen compounds were measured in sediment and found to be below laboratory detection limits. This is expected as nitrogen based compounds would not typically adsorb to soil in significant quantities. Localised effects due to the nutrients in groundwater discharges (e.g. algal growth at seepage points) are discussed in Section 7.

The sandy gravel that replaced the excavated material should have been clean immediately after being placed. It is apparent that this material has been re-contaminated as samples from within the backfill had DDX concentrations of between 0.4 and 6.4 mg/kg (average of 1.7 mg/kg).

When considering the overall contaminant mass removal, it should be noted that these concentrations are not necessarily representative of backfill mass as a whole. This is because the laboratory results are based on the passing 2 mm fraction and do not take into account the mass of material above 2 mm. The analytical results are therefore likely to be more representative of fine sediment that has migrated into the coarse fill material from outside the excavation. Without a particle size distribution for the sandy gravel that was used as backfill, it is difficult to quantify this effect. However, based on the description and appearance of the fill material, a reasonable estimate of the percentage of material less than 2 mm might be 10%.

If this was the case, the effective DDX concentrations in the fill material are actually lower by a factor of 10 (i.e. with an average of about 0.2 mg/kg). How this might be compared with earlier sediment samples is not known. However, where previous samples were predominantly of mud, rather than of sandy gravel, the complete particle size range of the sample would have been analysed and therefore the sample would be more closely representing the whole sediment.

The current beach, mainly gravel with sediment between larger particles on the surface, provides a different habitat to the original foreshore. This is evidenced by the change from the mud snail to the top shell as the dominant snail species. The change in biota may mean that the residual contamination is less important for the new species, being surface dwellers rather than burrowers, but more study would be required to ascertain this.

6.7.1.3 East Recontamination Mechanism

There is not enough information to determine the mechanism of re-contamination of the remedial fill material. However, some of the potential causes are:

• discharge of contaminated sediment via runoff from the site during the remediation works (this appears to have been confirmed as a mechanism in the west, as discussed in Section 1.9.3 below). This includes discharges during the remediation of the surge chamber. Given the eastern foreshore remediation occurred early in the overall remediation project, there was a greater potential for discharges from the site than if the remediation had occurred later;

• discharge of stormwater from the historic site stormwater network;

• migration of surface sediment from outside the excavation extent;

• migration of contaminated fines from the underlying sediment because of grain-size incompatibility between fine underlying sediments and the coarser gravel backfill. Strong currents, tidal fluctuations and wave action may have caused sufficient water movement within the gravel to cause contaminated fines to migrate to the body of the replacement gravel and perhaps the surface over time. This is why a graded sand or geotextile filter is typically placed between soils of different grain sizes in locations where they are exposed to water flows; and

• the potential mixing of the fill with the underlying sediment by the root-raking process (see Section 1.5.3). The likely significance of this is difficult to quantify with the available information. However, if substantial mixing of underlying material did in fact occur, this alone could have been responsible for the subsequent elevated concentrations.

Contamination of sediment via groundwater discharge is not likely to be significant. It is possible to use the groundwater flux estimates discussed in Section 7 and typical average contaminant concentrations in on-site groundwater to estimate the potential rate of discharge of OCP contamination to the marine sediments. It is immediately apparent that the rate of re-contamination of sediment through this mechanism would be slow, even if it assumed that all the dissolved contamination is adsorbed to a thin layer of sediment over the foreshore, which is unlikely. The re-contamination of the marine excavation occurred over a short period with a rapid increase in contaminant concentrations. If the contaminated groundwater discharged in a relatively localised area and concentrated in the near surface sediments, it is possible that discharges over the long term (years) could increase sediment concentrations measurably. This would become more significant if the adsorbed contamination was concentrated in the surface of the sediment. However, the level of increase would still be relatively small compared to the residual concentrations that are already present at depth.

As discussed in Section 7, the hydrogeological model for the site is uncertain and any such predictions of contaminant flux are also uncertain. Direct measurement of the sediment quality over time is preferable.

Of the potential mechanisms outlined above, the first has been effectively mitigated by the placement of a capping layer of clean soil across the site. Consequently, one of the key potential sources of sediment contamination has been removed i.e. via sediment runoff from the site. This source removal should be recognised as a key facet of the overall sediment remediation. The potential concentrations in sediment discharging from the site while the site was operational and during the remediation works were significantly higher than the current potential sources and the residual contamination in the remediated area.

In a similar manner, the historic discharge from the site stormwater system has been removed as a potential source of contamination of the sediment. The stormwater system was probably effectively cut off before the east sediment remediation, and so was unlikely to be a significant source of re-contamination of the excavation. However, prior to the surge chamber being removed, some stormwater discharges may have still been occurring.

The fact that the key contamination sources from the land have been removed appears to be consistent with the most recent sediment monitoring data from TDC, which indicates that the top 1-2 cm of sediment is improving in quality, probably from deposition of ‘cleaner’ sediment from elsewhere in the channel. For example, the DDX concentrations in a sample from adjacent to a seep at the top of the beach have reduced from 2.4 mg/kg in 2007 to 0.3 mg/kg in the February 2009 monitoring round. The DDX concentrations in shallow sediment further down the beach are an order of magnitude lower. A similar reducing pattern is apparent in ADL concentrations, although the concentrations are much lower than the DDX concentrations. Dieldrin concentrations in sediment samples from the most recent monitoring round ranged from 0.05 mg/kg to less than the laboratory detection limit of 0.001 mg/kg.

The elevated concentrations of DDX compounds outside the excavation remain a potential source of contamination. In time, equilibrium between the surrounding area and the remediated area is expected to occur. Much of this contamination is at depth or beneath gravel armouring and is less likely to mobilise. In addition, as the Mapua Channel is a high energy environment, sediment deposited adjacent to the site is likely to be sourced from a relatively wide area (in addition to sediment runoff from the site). Further away from the site, contaminant concentrations are much lower, potentially resulting in a lower ‘average’ contaminant concentrations in the sediment deposited on the eastern foreshore.

As discussed, the mechanics of the original recontamination are not well understood, but if it is assumed they are mostly related to the activities on the site or the way the remediation was performed (including coarse gravel over finer sediments), it is probable that these mechanisms are no longer occurring or are much reduced. The possible movement of fines from the contaminated sediment at the base of the excavation, if it has occurred, will reach equilibrium in time (and may have done so), as the pore spaces in the sandy gravel are filled with finer sand and silt, forming a natural filter preventing further migration from this source.

The remaining mechanism is a general redistribution of surface sediments (including some ‘clean’ site runoff) resulting in a gradual averaging out of concentrations. This will be less than the original foreshore surface concentrations but may never be as low as the SACs in the medium term because the average of the surrounding area is too high. Over time other processes will reduce concentrations through the gradual breakdown of the organic contaminants. With half lives of years, this will be a gradual process.

Continuing direct measurement of the sediment quality is recommended. The length of the surface sediment takes to reach equilibrium with the surrounding area is not known, but it is expected to be a few to several years. Slow natural degradation will then occur, reducing concentrations over time. Calculating the length of time this slow degradation will take to reduce concentrations to the SACs would be a useful exercise to carry out before contemplating any further remedial action.

6.7.2 Remediation to Extent Practicable – East

In a broad context, remediation to the extent practicable has been achieved for the marine sediments in the east. This overall conclusion does not imply that the remediation could not have been carried out more efficiently and achieved a better result for the replacement gravels. The exercise could be repeated and a better outcome could be achieved for the bulk of the sediment, albeit at a cost, given the risk of re-contamination from the site is much lower. The sediment is not so contaminated that it would create a major disposal problem (as it meets residential criteria), although it would have to be handled carefully.

However, re-deposited surface sediment will still likely be at greater concentrations than the SACs, reflecting the surrounding concentrations. In addition, the benefits of further remediation are likely to be outweighed by the ‘costs’ such as additional disruption to the current habitat, potential impacts on the wider environment and financial cost, for an outcome at the surface that will not be very different from the current outcome. There would likely be further delay to using FCC East.

The physical constraints are recognised as a key factor relating to whether remediation to the extent practicable has been achieved. Working beneath water is difficult and risks unacceptable effects such as sediment discharge to the estuary. It would be possible to create work areas isolated from the tidal movement, potentially beyond the low tide mark. However, this would be expensive. In addition, water-laden sediment by its nature is difficult to work with.

During the remediation works, the site was used to dewater and dispose of the sediment. For future works, either part of the FCC East site would have to be re-established by stripping the capping layer off over a sufficient area and establishing a bunded working area, or an alternative area would need to be found and secure transport used to prevent discharges during transport. It is inevitable that at least a staging area would have to be established on FCC East. While these things are quite possible, re-excavation would be a more expensive proposition than when carried out as part of the larger works. This reduces the practicality of any further remediation.

Elevated OCP concentrations (mainly DDX) remain in shallow sediment at locations in the wider area that are not practical to remediate, due to physical and cost constraints discussed above. These include locations beneath the nearby wharf and up to 100 m north of the site, adjacent to Mapua Channel. If it is accepted that the entire area of sediment contamination cannot be remediated, this brings into question the benefit of remediating only part of the area.

In addition, the limited data available appears to show that the shallow sediment quality is improving. It may be that the improvement in the chemical quality of the marine habitat will continue without further remediation. Additional data would be required to confirm this conclusion. The removal of the site itself as an ongoing source of contaminated sediment is a key aspect of the sediment remediation, that is, by preventing further runoff of contaminated sediment. This alone has likely resulted in an improvement in surface sediment quality and will allow natural attenuation to slowly improve the situation further. If the improvement in surface sediment quality is sustained, the residual sediment contamination at depth assumes a lesser importance and further remediation of the deeper residual contamination becomes harder to justify. The vertical distribution of contamination and how this might relate to biota at depth in the sediment is not currently well defined. It is therefore difficult to estimate effects of deeper contamination. If there are no receptors affected by the deeper contamination, its significance is reduced.

As discussed in Section 1.3, the residual OCP concentrations in the sediment are unlikely to represent a risk to the health of recreational users through direct contact. The results are typically well below equivalent human health guidelines (particularly for surface sediment) and exposure would be expected to be intermittent. Consequently, additional remediation is not warranted for this exposure pathway.

Based on the recent monitoring results, no further remediation is warranted due to risks associated with the consumption of seafood in the east. The sediment SAC were based on the consumption of seafood by humans, in particular of the mud snail (amphibola crenata). However, these snails are no longer present and the new species is unlikely to be consumed. In any case, based on recent results (TDC, 2009b), the concentrations of OCPs in the flesh of the snails are not at a level of concern for human health even if the snails were consumed. There are other seafood sources in the area with the potential to bioaccumulate. However, a previous study concluded that snails were the most representative bio-indicators for the site (Landcare Research, 2002). For example, crabs, oysters and cockles accumulated lower levels of OCP contamination relative to snails. Further monitoring is justified to confirm that bioaccumulation stays at the present levels.

The average concentrations of DDX in the vicinity of the remedial excavation are well in excess of the ISQG-High marine sediment guidelines from ANZECC (2000), even if an allowance is made for the effect of total organic carbon. This suggests that some form of adverse effects on marine ecosystems is likely. However, similar concentrations exist in the wider area of Mapua Channel. It is therefore difficult to justify further remediating one part of the area without addressing the other. In addition, the available data appear to show that shallow sediment quality is improving in any case. Additional monitoring of the marine ecosystem is warranted to benchmark conditions and enable future improvements (or otherwise) to be gauged.

It should be noted that the habitat on the eastern foreshore has been substantially modified by physical changes imposed by the remediation itself. Where mud snails were previously present, these have been replaced by a mudflat top shell. This is already a significant change to the local marine ecosystem which should be recognised when assessing other effects.

The derivation of the SACs also considered effects of bioaccumulation at higher trophic levels based on a literature review of likely bioaccumulation factors (Egis, 2001). Given that the SACs were not achieved, higher trophic levels may now be at risk. A revisiting of the original study using site data could well show that the original study was too conservative. If so, given that food consumption does not currently appear to be a risk, the SACs could be revised upwards.

6.7.3 Western Foreshore

6.7.3.1 Mass of contamination removed

In a similar calculation to the east, the mass of DDX compounds removed by the west marine excavation is estimated to be in the order of 5 kg. For the west, there is slightly more data indicating the possible extent of the DDX contamination in the Waimea Inlet. Using the available data, it is estimated that in the order of 3.3 kg remains in the tidal mud flat portion of the estuary in the top 0.25 m, beyond the remedial excavation extents. This estimate is uncertain as it is based on limited data. Further contamination probably exists at greater depth, but this is less significant for the marine environment. In addition, the calculation does not include residual contamination in the rushes adjacent to the creek and at depth in the swale. These sources could also include significant quantities of DDX compounds, albeit in a potentially less ‘available’ location than the surface sediments on the tidal flats. However, the calculation indicates that approximately 60% of the available surface contamination was removed.

While this is less than the 90% suggested by the Site Auditor, it represents the readily accessible contamination. Each additional 10% removal is increasingly difficult and increasingly expensive, with the law of diminishing returns operating. On that basis, it is reasonable to conclude a significant mass of contaminant was removed.

6.7.3.2 West Current Status

The sediment quality in the vicinity of the western foreshore has been characterised to a slightly better level than the eastern foreshore. Initial post remediation sampling indicated that DDX concentrations above 1 mg/kg were present at the extents of the remedial excavation. However, the vast majority of these were located in the stormwater drain (‘the creek’), where approximately 15 of about 50 validation samples had concentrations above 1 mg/kg. Three creek validation samples were above 15 mg/kg, with a maximum of 82 mg/kg detected on the north-western bank of the creek.

On the foreshore itself, the initial post-remediation samples were typically well below 1 mg/kg and often below 0.1 mg/kg. Foreshore samples taken at various times in 2007 showed DDX concentrations in shallow sediment of between 0.01 mg/kg and 24 mg/kg, with concentrations typically above 1 mg/kg. These samples were taken over a year after the remediation but while works on the adjacent FCC Landfill were still occurring, potentially indicating some recontamination from site runoff. This is consistent with the apparent subsequent improvement in surface sediment quality after the site was capped with clean soil and grassed (see below).

DDX concentrations in the wider bay are typically well below 1 mg/kg, with an average in the order of 0.15 mg/kg (see Section 1.8), although still 10 or more times above the SAC.

The residual ADL concentrations on the western foreshore are much lower than the DDX concentrations, but still elevated. All ADL concentrations were below 1 mg/kg, with a maximum of 0.49 mg/kg detected. Many of the samples taken from the remediation excavation extents had concentrations of less than 0.01 mg/kg. However, the various sampling events in 2007 indicated concentrations in shallow sediment were typically in the order of 0.1 mg/kg. These results are likely to represent contamination from site runoff.

As with the east, the most recent sediment monitoring data from TDC appears to indicate that the top 1-2 cm of sediment is improving in quality. For example, the DDX concentrations have reduced from typically well above 1 mg/kg in 2007 to generally less than 0.3 mg/kg in the February 2009 monitoring round (a concentration of 0.6 mg/kg was the exception to this). These concentrations are now closer to the approximate average concentration in the adjacent bay (0.15 mg/kg). This would be consistent with movement of shallow sediment from the bay into the remediated area.

A similar reducing pattern is apparent in ADL concentrations, although the concentrations are much lower than the DDX concentrations. Dieldrin concentrations in sediment samples from the most recent monitoring round ranged from 0.02 mg/kg to 0.004 mg/kg.

6.7.3.3 Recontamination Mechanism

The sampling results for the western foreshore suggest similar mechanisms of re-contamination of sediment as the east, that is:

• discharge from site during the remediation works;

• migration of sediment from outside the excavation; and

• migration of sediment from the base of the excavation into the backfill.

As noted previously, contamination from groundwater is not expected to be significant, and certainly not as rapid as the rate at which the initial re-contamination occurred.

There are a number of differences in the situation in the west. Anecdotal reports suggest quite strongly that site discharges did cause at least some of the re‑contamination of the remedial excavation, whereas this is less certain for the east. In addition, The DDX concentrations detected in shallow sediment shortly after the remediation were greater than those present in the other main potential source i.e. the adjacent bay. As with the east, the site as a source of sediment contamination has now effectively been removed with the completion of the remediation and establishment of grass on the site.

In the west, the concentrations at the base of the foreshore excavation (i.e. excluding the creek) were typically much lower than in the east. Consequently, it is much less likely that significant re-contamination occurred from the base of the excavation. In any case, as discussed above, this form of migration is more likely to have reached equilibrium by now.

There are also elevated concentrations (relative to the marine SAC) of DDX present in the sediment at the base of creek excavation. These residual DDX concentrations are a potential ongoing source of contamination to the sediment in the bay. However, it is difficult to predict whether this is a significant mechanism relative to other potential re-contamination mechanisms. The creek was covered with a layer of clean gravel and the banks of the creek are now heavily vegetated. Both these factors would tend to reduce the potential for sediment to mobilise from the creek bed.

In May 2007, CH2M HILL collected samples from the creek bed. Samples from within and immediately downstream of the area with the highest residual contamination returned low DDX concentrations. This suggests that the residual contamination along the creek is not particularly mobile, although the data are limited.

This leaves the wider bay as the main potential source of ongoing contamination, but not the predominant source of the initial re-contamination. Previous investigations have shown that sediment with elevated DDX concentrations extends up to 250 m to the south of the western foreshore. It is probable that this is the main ongoing source of sediment contamination for the foreshore. Consequently, it is likely that sediment concentrations close to the beach will reach equilibrium with concentrations in the wider estuary, before then slowly degrading.

It is difficult to estimate the proportion of the overall estuary sediment load that these sources contribute, and hence the likely equilibrium that will be reached. The most recent sediment monitoring round reported DDX concentrations ranging from 0.1 to 0.6 mg/kg. Similar concentrations were present in the wider bay in 2007. Further data would be required to confirm trends in shallow sediment concentrations, but ultimately it is expected that the surface sediment (0 – 50 mm) will reach an average concentration similar to the wider bay, i.e. 0.15 mg/kg ¹⁷. Natural degradation process will also be occurring concurrently, but at much slower rates. As for the east, further study would be required to estimate the rates of natural degradation.

6.7.4 Remediation to the Extent Practicable – West

Remediation to the extent practicable of the marine sediments has broadly been achieved in the west, for the same basic reasons outlined for the eastern sediments; specifically, that it is not possible to remediate the surface sediments to a higher standard than the surrounding surface sediments.

A potential exception to this is the creek area where re-sedimentation from the wider bay is less likely and the residual contamination is a small volume and quite accessible. However, in this case, additional remediation is not warranted as the residual DDX concentrations are not likely to be currently posing a significant risk. The edges of the creek are heavily vegetated (and typically above water), with little potential for mobilisation of or direct contact with the sediment. This appears to be supported by recent sediment results from the creek which were lower than other areas (CH2M HILL, 2007). However, a check should be made that the flood flows in the creek are not likely to be so high as to cause erosion and the Site Management Plan should include measures to ensure that the creek and banks are not excavated without controls (e.g. during some future drainage improvement works) and are protected from future erosion. Sampling of creek-bed sediments should continue on an annual basis to check that contaminated sediment is not being remobilised. Such monitoring could cease if a lack of remobilisation is confirmed.

Data from 2007 showed that elevated concentrations of DDX compounds existed across much of the tidal flat area to the west of Grossi Point, extending up to 300 m south of the western foreshore. It is not practical to remediate this entire area due to cost, difficulties in handling and disposing of the large quantities of sediment, and likely detrimental effects on the estuary ecosystem. Further remediation of the foreshore area in isolation is therefore not justified.

As with the east, the removal of the site as a source of sediment contamination is a key aspect of the remediation. In the west, the sediment sampling history suggests that site runoff was a major mechanism of re-contamination of the foreshore. It is now likely that the contaminant concentrations in the shallow sediment will reach approximate equilibrium with the concentrations in the wider bay (and ultimately the estuary). Even at the concentrations of the wider bay, some form of impact on the marine ecosystem would be expected, based on a comparison with the ANZECC (2000) marine sediment guidelines. As with the east, additional monitoring of the marine ecosystem is warranted to benchmark conditions and enable future improvements (or otherwise) to be gauged.

The available information indicates that effects on other potential receptors from the residual sediment contamination are not likely to be significant, except possibly for human consumption of seafood. In the case of risks to human health via seafood consumption there is uncertainty as the dataset is small. However, the most recent data indicate that average concentrations in snail flesh are below levels that are likely to cause a significant human health risk. The New Zealand Food Safety Authority provided advice to TDC (TDC, 2008a) that safe average daily intakes would not be exceeded if seafood concentrations were less than 21 mg/kg wet weight for DDX and 0.2 mg/kg wet weight for the sum of aldrin and dieldrin. The most recent DDX concentration of 3.5 mg/kg is well below the guideline and the dieldrin/aldrin concentration is essentially at the criterion of 0.2 mg/kg, indicating a significant risk is currently unlikely. Further monitoring is required to confirm this conclusion as the sample size is too small to be certain.

6.7.5 Recommendations – Marine Areas

As discussed previously, the tidal dilution is such that effects in the wider estuary are unlikely to be significant, particularly in the Mapua Channel. However, there will be some localised effects on the marine ecosystem from the residual OCP contamination in sediment, and from nutrients in the groundwater discharge on the foreshores. The available data indicate that human health is not likely to be at significant risk from the residual contamination in the marine environment (either from direct contact or ingestion of seafood). In addition, recent data appears to indicate that the localised effects are likely to be reducing as surface sediment quality appears to be improving.

The shallow sediment has broadly been remediated to the extent practicable, essentially as it is not possible to remediate the surface sediments to a higher standard than the surrounding surface sediments. The remediation of deeper sediments, while theoretically “sustainable”, would be hard to justify on an effects basis. Determining what level of effects on the marine ecosystem is acceptable is a decision for stakeholders such as TDC and MfE. However, if it is accepted that further remediation of the shallow sediment is not practicable, it must also be accepted that some localised impacts on the marine ecosystem are unavoidable.

The following recommendations are made for additional investigation and monitoring associated with the marine areas:

Biota and Sediment Monitoring

The current annual monitoring of sediment and biota by TDC should be continued and expanded. The aim of the monitoring will be to:

• confirm OCP concentrations in snails (as appropriate bio-indicators) remain below levels that might present an unacceptable risk to human health;

• to confirm apparent improving trends in the chemical quality of shallow sediment using a larger sample set; and

• to provide additional information on localised effects of nutrients in groundwater discharges on the foreshores (see Section 1.19.2).

Prior to undertaking the next monitoring round, we recommend a review is undertaken by an appropriately qualified person to confirm that monitoring snails on the eastern foreshore is the most appropriate method of assessing risk via seafood consumption. The new species of snail on the eastern foreshore (mudflat top shell) appear to bioaccumulate less than the original mud snail. Consequently, the new snails may not be an effective bio-indicator. The review should assess the previous reports on the subject, including that by Landcare Research (2002) and take into account recent monitoring data and the likely site use. Consideration should be given to the need for confirmatory sampling of other biota.

Monitoring should be undertaken as follows:

• the annual monitoring frequency should be continued, with the monitoring scope reviewed after two additional monitoring rounds;

• in addition to the current monitoring locations the following sediment sampling locations should be considered:

• three additional locations parallel to the western foreshore, approximately 20 m from the foreshore edge. The locations should be evenly spaced along the foreshore;

• two additional locations parallel to the eastern foreshore, approximately 5 m from the base of the sea wall. The locations should be evenly spaced between the current sampling transect and either end of the foreshore.

• at each sediment sampling location, samples from 0 – 0.02 m and from 0.02 – 0.10 m should be collected. A sediment corer with a core extruder should be used to ensure accurate sample depths. Each sample should be analysed for DDX and ADL;

• the snail sampling should continue as previously, unless otherwise indicated by the review on biota sampling outlined above;

• total organic carbon (TOC) should be measured in each sediment sample in the first monitoring round;

• a particle size distribution should be undertaken on 50% of the sediment samples in the first monitoring round;

• detailed field and photographic records should be kept of all observations, e.g. sediment colour, number/size of snails, etc;

• a photographic and written record should be maintained of areas of algal growth. The photos should be taken from the same perspective to enable comparison between monitoring events (see Section 1.19.2);

• at each sediment sampling location, samples from 0 – 0.02 m and from 0.02 – 0.10 m should be collected. A sediment corer with a core extruder should be used to ensure accurate sample depths. Each sample should be analysed for DDX and ADL;

• the snail sampling should continue as previously, unless otherwise indicated by the review on biota sampling outlined above;

• total organic carbon (TOC) should be measured in each sediment sample in the first monitoring round;

• a particle size distribution should be undertaken on 50% of the sediment samples in the first monitoring round;

• detailed field and photographic records should be kept of all observations, e.g. sediment colour, number/size of snails, etc;

• a photographic and written record should be maintained of areas of algal growth. The photos should be taken from the same perspective to enable comparison between monitoring events (see Section 1.19.2);

Marine Ecosystem Health and Bio-diversity Monitoring

Effects on the marine ecosystem are likely given the incomplete remediation of the foreshores. Such effects have not been gauged to this point. It is recommended that investigation of the new foreshore ecosystem be carried out to benchmark conditions and provide a basis for assessing expected improvements in the foreshore sediments into the future.

Determining the biological health of the marine environment close to the site relative to before the remediation is problematic as no benchmark studies were completed prior to the remediation with which results can be compared. However, a controlled study may be achievable by locating similar habitats to the east and west foreshores elsewhere in the inlet and making comparisons. There are a number of survey techniques which can be used to assess ecosystem health and bio-diversity. Sediments at the chosen sites would need to be sampled for OCPs.